530 
APPENDIX. 
[Errors in variation. 
dipped about 13° only, I had some observations which countenance this sup- 
position, though taken at too great a distance to afford a satisfactory proof. 
Trinidad S. S. W. 13 leagues, corrected variation 3° 58' west. 
S. S. E. 6 leagues, - -- -- -4 15 
N. E. by E. 15 leagues, - - - - 1 50 
The attraction of land has here been supposed to be so far uniform, 
as that, in high dips of the needle, the upper part has the same kind of at- 
traction as the magnetic pole of its hemisphere ; and that near the mag- 
netic equator, the north end of the land attracts the south point of the 
compass, and vice versa; but it must be evident, that not only should there 
be a regular gradation from one to the other, as the dip increases or dimi- 
nishes, but also that local discordances may take place in both cases, where, 
instead of one extended magnetic substance pervading the whole land, 
there are many detached masses, veins, or blocks. Each one of these will 
possess anorih and a south polarity; and contrary attractions may there- 
fore be found at different degrees of elevation and in short intervals of space, 
without the supposition of the general attraction in land being thereby 
overturned. 
That even small differences in elevation may produce a change in 
the magnetic needle appears from some instances in this voyage, where 
observations were taken on shore, either in the same spot or a few feet from 
it, with the ship’s azimuth compasses and with a theodolite. The com- 
passes stood on the ground, the theodolite upon legs about four and a half 
feet high, and the variations were as under : 
In K. G’s Sound, from three compasses 6° 23', from theodolite 8° 17' W. 
Lucky Bay, 2 35, - - - - 0 30 W. 
Kanguroo I., from one compass 2 58, - - - - 5 48 E. 
Some part of these differences might arise from erroneous construction of 
the instruments, but only a small part ; for they were scarcely sensible in 
other cases where no hills existed in the neighbourhood, or where the 
theodolite was placed on the highest land. It is to be remarked, that the 
compasses come nearer to what appeared to be the true variation than did 
the theodolite ; which I should attribute to the attraction being more equal 
all round, upon the instrument placed on the earth, and to the theodolite 
being influenced by the neighbouring hills. . - 
